Edit model card

distilbert-base-uncased-finetuned-emotion

This model is a fine-tuned version of distilbert-base-uncased on the emotion dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1495
  • Accuracy: 0.9385
  • F1: 0.9383

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 64
  • eval_batch_size: 64
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss Accuracy F1
0.1739 1.0 250 0.1827 0.931 0.9302
0.1176 2.0 500 0.1567 0.9325 0.9326
0.0994 3.0 750 0.1555 0.9385 0.9389
0.08 4.0 1000 0.1496 0.9445 0.9443
0.0654 5.0 1250 0.1495 0.9385 0.9383

Framework versions

  • Transformers 4.16.0.dev0
  • Pytorch 1.10.0+cu111
  • Datasets 1.18.0
  • Tokenizers 0.10.3
Downloads last month
13

Dataset used to train Fengkai/distilbert-base-uncased-finetuned-emotion

Evaluation results